d8ae766099
Add a bunch of .round()s before conversions
...
Previously this was leading to inconsistencies when using f32 v.s. f64
2020-12-17 15:09:53 -08:00
75fde43aa7
Get another 25% perf bump by optimizing material layouts
2020-12-16 14:30:52 -08:00
41572f5285
Optimize GenericMaterial layout
...
Shaving off 2 floats, i.e. 16 bytes. Brings a 3-4% overall perf
increase. Should be possible to shave another 16 bytes.
2020-12-16 14:04:30 -08:00
737ae1b872
Add diagnostics binary for showing things like material parameters
2020-12-16 13:53:39 -08:00
c9b4ab5b6e
More reactive z navigation
2020-12-16 00:10:16 -08:00
6362b5a60e
parallelize render_scalar_field
2020-12-15 14:19:29 -08:00
b3aba7658b
Add a magnetic loop measurement
2020-12-14 21:54:33 -08:00
cba2e3c3d2
Change up some measurement/rendering details
2020-12-14 21:44:07 -08:00
e105819461
CurlStimulus can now handle negative magnitudes
2020-12-08 22:19:06 -08:00
f5f90a7a15
Fix a bunch of warnings
2020-12-07 22:55:24 -08:00
e0056d8c85
Make the Driver able to handle any Material
2020-12-07 22:51:00 -08:00
f1c9334322
Don't recalculate inv_feature_size in every work unit
...
This shaves nearly 1% off the total simulation time
2020-12-07 22:05:20 -08:00
82b1eea353
Add a way to measure the current directed around a closed loop
2020-12-07 09:16:07 -08:00
d9176cdf06
Non-uniform current stimuli can be specified easier
2020-12-06 20:56:34 -08:00
d619383fb9
Make Current measurement serializable
...
Note that this REALLY slows it down, since it forces the Region to be
Boxed.
It might make sense at some point to consider introducing a RegionBitmap
type. Maybe keep the original Region type -- for serializing -- but
cache a Bitmap version where perf matters more.
2020-12-06 19:54:02 -08:00
8d4a3a91fe
Fix broken tests + misc cleanup
...
Move all the examples which don't compile to the `old/` directory.
2020-12-06 13:31:13 -08:00
28c734a4c5
Add a Current stimulus to the wrapped torus
2020-11-29 21:52:52 -08:00
1f44605714
Add an example for a "wrapped torus": that is, a ferromagnet torus with wire loops around it
...
Still missing:
- stimulus to inject a current onto one of the wire loops
- measurement to measure current on the other wire loop
2020-11-29 18:25:34 -08:00
ed86aff504
measurements are now rendered alongside the terminal view
2020-11-29 12:27:52 -08:00
2c042810d8
Implement a 'SerializerRenderer' output format
...
It's CBOR, which seems to pack not so efficiently (250*250*20 * 12
floats takes 400 MB per frame, whereas this many doubles could pack into
120 MB). serde_cbor also seems to be extremely slow: taking multiple
minutes per frame. This could be parallelized to be less problematic
though (it already is -- just bump the parallelism).
Might be worth testing other formats.
2020-11-27 22:11:53 -08:00
9d15e126a7
Cleanup
...
AKA I made these changes a month ago but forgot to commit them and don't
really want to figure out what they did.
2020-11-27 21:22:42 -08:00
1a9093315a
Basic support for plotly backend
...
NOTE: this requires you to clone `plotly`, copy the
`plotly_kaleido` folder to be adjacent to this one, and then run `cargo
build --release` from within that plotly_kaleido folder.
2020-10-09 21:51:34 -07:00
f2ba6fdd1b
Troubleshoot toroid2.5d
2020-10-09 20:41:38 -07:00
4ac48bd3f1
Minor cleanup to Render size
2020-10-03 22:13:59 -07:00
0642f52172
Fix some issues related to the Courant number
...
step_h was using the wrong timestep; fixing it caused update issues
based on the Courant number being 1.0, i.e. too large. Reducing it to
\sqrt{3}/3 fixes stability issues. Unfortunately, a non-Unity Courant
value prevents me from using the ideal Absorbing Boundary Condition I
had just implemented, so that's been disabled.
2020-10-03 19:44:30 -07:00
5c8c2f6c76
Fix division by zero in UPML boundary
2020-10-03 13:58:44 -07:00
55d6171325
Add units to the different Vec3 "types" (Meters and Index, so far)
2020-09-26 17:05:15 -07:00
00f22c95d0
WIP Stimulus system
2020-09-26 15:50:34 -07:00
9a0f778db5
UPML boundary now applies to z dimension
2020-09-26 15:13:30 -07:00
d39db969b5
Consolidate Vec/Coord into the geom submodule
...
Also add the missing coord.rs to version control...
It should have been checked in earlier
2020-09-26 14:09:41 -07:00
452141dd0d
Change meas.rs to use meter coordinates instead of pixel coordinates
2020-09-26 14:01:10 -07:00
729f9f77a6
Make the Region trait 3d-ready
2020-09-26 13:39:34 -07:00
bef423ce38
Rename Point -> Vec2 to distinguish it clearly from any 3d object
2020-09-26 13:25:49 -07:00
fa0e30a2a6
Make Coord 3d
2020-09-25 21:04:16 -07:00
6d77c85179
Abstract coordinates behind a Coord type
...
This will allow adding another dimension to the coordinate, more easily.
2020-09-22 11:23:58 -07:00
2549ac9940
step_h uses optional left/down nodes.
...
This will be handy for implementing 3d, because the world with depth=1
(i.e. no neighboring xy planes) will behave the same as a 2d simulation.
2020-09-22 08:56:41 -07:00
253ffe585f
Fix current measurement
2020-09-20 11:53:05 -07:00
d713c79bbd
Improve the UPML parameters by properly summing the axis conductivities in the corners
2020-09-19 20:10:10 -07:00
f6e4ed8cb9
Add a UPML boundary condition (needs improving in the corners, though)
2020-09-19 20:02:11 -07:00
ae93101676
Allow conductivity to be specified as a vector
...
This is necessary to implement UPML boundary conditions
2020-09-19 19:33:11 -07:00
a7899d9be1
Add a toroid example where the B field is actually directed circularly inside the toroid
...
This is a more accurate simulation of a core memory.
Presently, trying to characterize the device: understand how the drive
current, and sizing, affects switching time.
2020-09-18 22:23:37 -07:00
f50dbf7d4e
Make the float depth compile-time configurable
2020-09-13 22:38:00 -07:00
d4334565a7
Add an example where the ferromagnet is a ring around the conductor
2020-09-13 18:33:31 -07:00
75008a6bd5
Consolidate duplicate 3r1 material definitions into mat::db module
2020-09-13 18:32:43 -07:00
53da11c874
Assorted minor aesthetic touchups
2020-09-13 17:19:44 -07:00
cb9f413710
Add energy measurement
2020-09-13 17:08:33 -07:00
0fc973b409
Toy around with some other setups.
...
It's weird (maybe), that the behavior of the ferromagnet is to _prevent_ H
from getting very large
2020-09-09 22:41:41 -07:00
4d8ef1c9f3
Add a measurement for magnetization
2020-09-07 21:58:07 -07:00
d611e822dd
Add a 'Measurement' concept, e.g. to display current at a point.
2020-09-07 20:14:41 -07:00
fbca8622df
Remove usize casts
2020-09-07 17:54:50 -07:00
